Sorry if doubly posted. Damn third world ISPs.
Ever been to the Netherlands? It's not the United States, believe it or not. Their judicial history and mindset is different than yours, or more accurately, what you perceive those of your country to be (I'm assuming from your solipsism that you're American). They don't consider the American '"separation of powers" thing' so pesky as it doesn't apply to them.
And if you think judges don't make new law in the U.S.A. and neither are they supposed to, you are sorrily mistaken, young v1. They are not supposed to use their powers of compulsion to harass people whose behavior they abhor into acquiescing to non-existent laws, but that's a very different thing.
You're half right. Every one of the, uncancellable, option trades had to be covered by a market maker in the option market on which it was sold. (A market maker is a firm that is always willing to buy at some price and sell at a slightly higher price, so that there is always a market.) The market maker doesn't just assume all the risk -- it covers it self by buying and selling the stock underlying the option, from the equity market, in this case nasdaq. These market makers, at least one of whom, per exchange, is obligated _always_ to provide a market for the stock, get their equity trades rolled back, also, even thought they calculated the price of the options based on the erroneous underlying -- if the stock is trading at $40 the right to sell it next month @ 45 is worth at least $5, a little more since there is time value -- it could go up and make more money, but it can exercise right now and get that $5 profit.
So the market makers are fucked because they sold all the options at bad prices, had the equity trades they used to offset the risk they are obligated to take removed, and now must scramble to price the options correctly, since the historical data used to price an option, i.e. volatility of the nderlying stock price is now hosed.
For every angle shooter, there was (at least) one situationally innocent player getting the same reaming.
Now, if only Watterson would get inspired to further the adventures of Calvin, there would be some ubiquity in the "Intellectual Section" of the daily fishwrap!
What is "ubiquity" supposed to mean in this +5 interesting comment?
the fact that this might have a positive result doesn't mean you should be "all for it". I'm "all for" the elimination of AIDS, but not by means of killing every HIV+ person in the world.
OK, I'll play weenie -- it compiles fine and warning -free if you use g++, because the c++ standard says that there is an implicit return of 0 if there is no explicit return from main (which also can no longer return void). Hence:
It is a (successful) attempt by copyright holders to frame the issue in emotionally-sensitive terms.
Well, while we're being technical, at least as early as 1755 "pirate" had come to mean "Pirate (Piraat)
n.f. [grieks, piratica, Lat. pirate Fr.] I. A fea-robber. II. Any robber; particularly a bookfeller who feizes the copies op other men."
There are many evils you can ascribe to the RIAA, perverting "piracy" isn't one of them.
Within 1 minute, I found that it's listing sans-serif fonts as serif, and serif fonts as sans-serif. Yikes.
Actually, it's allowing you to select _any_ font you want to be displayed when the page author has suggested serif, it's not saying "these are serif fonts, pick one". Therefore, both dropdowns contain all installed fonts. It's a feature, not a bug.
I will admit, however, that toolbar weirdness is probably not a feature.
They are obviously planning on doing some insider trading, selling out the investors when the stock is at it's peak, long before the inevitable crash.
The lawsuit is completely arrogant and bogus. Check.
What there intentions are is not obvious at all, and attributing insider trading motives to them is along the same lines as "couldn't implement technology X without stealing it from UNIX". You don't know, you're speculating and the results of your speculation are not "obvious" at all.
Besides which, the term insider trading would only apply if the trading was done based on non-public information, in which case, it would certainly not be obvious to the/. crowd, at least those of us not on the SCO board of directors. Obviously.
The OS X kernel is, as I understand it, has both mach part and bsd part, and they both have access to the same memory, for the sake of speed of system calls, so it's not a microkernel, it just incorporates technology from a microkernel project.
Unless i'm wrong, in which case I'm sure i'll here about it.
Every single track on his last disc was licensed to corporate america to pimp, cars, Tvs, whatever.
every single one.
now he wants me to run out and spend 20 dollars on his critically panned shite, when he's planning on ramming it down my throat anyway every time I peep the Simpsons? So he wants me to pay to listen to it, then have someone else pay him to force me to listen to it.
Call me Ishmael, but i'd like to hit him w/ a harpoon my own self.
Do you mean to suggest that you wouldn't have offered "IBM Thinkpad" to Apple to pimp powerbooks if IBM had told you to get bent when offered your dubious service?
load of crap. they may come close to being speed-equivalent if you write c++ constrained to the language features java supports. That is, you use heterogeneous container classes and downcast all the time, don't use templates or multiple inheritance, use threads where there is no good reason to, and whip expensive exceptions around like you're getting a commision on them.
I did say *unoptimized -- I know that a properly coded and optimized application written in in C++ can easily outperform a properly coded and optimized Java application. My point was that Java is *fast enough*... given that no one has time in their development schedule to completely optimize anything, and no one has enough competent programmers to code it that way, even if they had enough time to do it and money to pay them.
Actually, you said 'an expert can still hand-optimize C++ to out perform Java', a position you're softening, but that's not even really the issue. The design of a C++ program can be such that unoptimized code almost necessarily runs faster than is possible with Java. Just look at using typesafe STL algorithms versus looping over java.util.Iterator and downcasting. Or using templates to effect compile time subclassing. BTW sorry I called your reasoning a load of crap. that was gratuitous.
Oh, and typical server applications?!? If you're writing server applications using a one thread-per-server-scheme, you are nowhere, and you are if you're writing java, cuz it doesn't know from multiplexing, at least in 1.3.
Ah... you're talking about using select() to code I/O handlers. I begin to see the problem -- you seem to think that a web server is a typical server application. I suppose I should have been more specific; I was referring to business applications -- ERP, DSS, HR, A/R, A/P, Inventory Management, etc -- the type of applications that businesses pay money to have developed, because they need them to run their businesses. These sorts of applications are rarely I/O bound, because they typically have no more than hundreds of users and spend most of their time waiting on database queries and user input. With this sort of application, the use of multiple threads to manage code execution is perfectly sensible, and results in very clean and easy to maintain code. You'll note that I was not advocating writing webservers or databases in Java -- but backend applications that serve webservers and use databases are excellent targets for Java development.
Fair enough, I don't know what type of code you are writing. And if Java serves your needs than use it; I use it myself because it is frequently the best tool. And FWIW, I was thinking more along the lines of/dev/poll or kqueue() than select()
Bash c#,.net, etc, all you want. But don't tell me an interpreted language with garbage collection and bullshit container classes and 1970's i/o is as fast as c++. I write networking code in both languages, and i'm not buying it.
Go back and read what I wrote -- I didn't claim that Java was as fast as C++ for *everything* -- just the sorts of applications I develop: business applications. Of course, if you continue to think of Java (and the JVM) as strictly an interpreted language, you're not keeping up with what the JIT compilers are doing these days, and you're completely ignoring the native-code compilers for Java.
Well, you said 'typical server applications' and you naturally think that means the kind you write, while I think it means the kind I write. I defer to you in determining what best suits your needs. I just get tired of people suggesting the JIT compilers or even native code producing Java compilers render C++, my personal fave, obsolete. Mostly just boosterism on my part.
Sorry for spelling: late, drunk.
Friends don't let friends post drunk.
The posting I can deal with. I just wish my friends would adopt some sort of stance on fat chicks.
Benchmarks have shown that for typical server applications, modern JVM/JIT combinations are as fast as unoptimized C++ code.
load of crap. they may come close to being speed-equivalent if you write c++ constrained to the language features java supports. That is, you use heterogeneous container classes and downcast all the time, don't use templates or multiple inheritance, use threads where there is no good reason to, and whip expensive exceptions around like you're getting a commision on them.
Oh, and typical server applications?!? If you're writing server applications using a one thread-per-server-scheme, you are nowhere, and you are if you're writing java, cuz it doesn't know from multiplexing, at least in 1.3.
Bash c#,.net, etc, all you want. But don't tell me an interpreted language with garbage collection and bullshit container classes and 1970's i/o is as fast as c++. I write networking code in both languages, and i'm not buying it. Sorry for spelling: late, drunk.
I work on Sun boxes all day long, because Sun's advertising and salesreps convinced someone in my company, who wouldn't know a line of code from a line of Goethe, that our company would be better off paying for Solaris than running Linux. That's a problem, and not one that's going to go away while Sun looks to make a profit.
All the king's horses and all the king's men (1 experienced Oracle DBA and 1 good Linux sysadmin) couldn't get DBD::Oracle working with 8i on Intel redHat 6.0 Linux at my shop. The make test failed loading a shared library, so I wrote a simple c prog just to load all the.so libs in a dir and it would load every.so file in the disk except any one in the oracle tree (Forget the error now, sorry). The kluging that followed was horrific.
Slashdot Mirror
The War Nerd on well, war, Scott Greenfield on (mostly criminal) law, and Ken White on law and privacy.
Any headline which ends in a question mark can be answered by the word no.
Sorry if doubly posted. Damn third world ISPs. Ever been to the Netherlands? It's not the United States, believe it or not. Their judicial history and mindset is different than yours, or more accurately, what you perceive those of your country to be (I'm assuming from your solipsism that you're American). They don't consider the American '"separation of powers" thing' so pesky as it doesn't apply to them. And if you think judges don't make new law in the U.S.A. and neither are they supposed to, you are sorrily mistaken, young v1. They are not supposed to use their powers of compulsion to harass people whose behavior they abhor into acquiescing to non-existent laws, but that's a very different thing.
You're half right. Every one of the, uncancellable, option trades had to be covered by a market maker in the option market on which it was sold. (A market maker is a firm that is always willing to buy at some price and sell at a slightly higher price, so that there is always a market.) The market maker doesn't just assume all the risk -- it covers it self by buying and selling the stock underlying the option, from the equity market, in this case nasdaq. These market makers, at least one of whom, per exchange, is obligated _always_ to provide a market for the stock, get their equity trades rolled back, also, even thought they calculated the price of the options based on the erroneous underlying -- if the stock is trading at $40 the right to sell it next month @ 45 is worth at least $5, a little more since there is time value -- it could go up and make more money, but it can exercise right now and get that $5 profit.
So the market makers are fucked because they sold all the options at bad prices, had the equity trades they used to offset the risk they are obligated to take removed, and now must scramble to price the options correctly, since the historical data used to price an option, i.e. volatility of the nderlying stock price is now hosed.
For every angle shooter, there was (at least) one situationally innocent player getting the same reaming.
What is "ubiquity" supposed to mean in this +5 interesting comment?
apologies if my analogy is overblown.
of course, if you consider it c++, you want to include cstdio and not stdio.h to be all namespace spiffy, but you knew that.
Well, while we're being technical, at least as early as 1755 "pirate" had come to mean "Pirate (Piraat) n.f. [grieks, piratica, Lat. pirate Fr.] I. A fea-robber. II. Any robber; particularly a bookfeller who feizes the copies op other men."
There are many evils you can ascribe to the RIAA, perverting "piracy" isn't one of them.
Rob, we're going to need you to turn in your key to the metaphor closet.
I think you mean increase signal to noise; if so, I agree wholeheartedly.
I'm from Brooklyn, i'm allowed to say "not for nothin'" unironically.
Actually, it's allowing you to select _any_ font you want to be displayed when the page author has suggested serif, it's not saying "these are serif fonts, pick one". Therefore, both dropdowns contain all installed fonts. It's a feature, not a bug.
I will admit, however, that toolbar weirdness is probably not a feature.
They are obviously planning on doing some insider trading, selling out the investors when the stock is at it's peak, long before the inevitable crash.
/. crowd, at least those of us not on the SCO board of directors. Obviously.
The lawsuit is completely arrogant and bogus. Check.
What there intentions are is not obvious at all, and attributing insider trading motives to them is along the same lines as "couldn't implement technology X without stealing it from UNIX". You don't know, you're speculating and the results of your speculation are not "obvious" at all. Besides which, the term insider trading would only apply if the trading was done based on non-public information, in which case, it would certainly not be obvious to the
The OS X kernel is, as I understand it, has both mach part and bsd part, and they both have access to the same memory, for the sake of speed of system calls, so it's not a microkernel, it just incorporates technology from a microkernel project. Unless i'm wrong, in which case I'm sure i'll here about it.
now he wants me to run out and spend 20 dollars on his critically panned shite, when he's planning on ramming it down my throat anyway every time I peep the Simpsons? So he wants me to pay to listen to it, then have someone else pay him to force me to listen to it.
Call me Ishmael, but i'd like to hit him w/ a harpoon my own self.
Do you mean to suggest that you wouldn't have offered "IBM Thinkpad" to Apple to pimp powerbooks if IBM had told you to get bent when offered your dubious service?
>Everyone could have an alterier motive. Or even an ulterior one.
The General public didn't see AI; they heard it sucked. They heard it from me.
I did say *unoptimized -- I know that a properly coded and optimized application written in in C++ can easily outperform a properly coded and optimized Java application. My point was that Java is *fast enough* ... given that no one has time in their development schedule to completely optimize anything, and no one has enough competent programmers to code it that way, even if they had enough time to do it and money to pay them.
Actually, you said 'an expert can still hand-optimize C++ to out perform Java', a position you're softening, but that's not even really the issue. The design of a C++ program can be such that unoptimized code almost necessarily runs faster than is possible with Java. Just look at using typesafe STL algorithms versus looping over java.util.Iterator and downcasting. Or using templates to effect compile time subclassing. BTW sorry I called your reasoning a load of crap. that was gratuitous.
Oh, and typical server applications?!? If you're writing server applications using a one thread-per-server-scheme, you are nowhere, and you are if you're writing java, cuz it doesn't know from multiplexing, at least in 1.3.
Ah ... you're talking about using select() to code I/O handlers. I begin to see the problem -- you seem to think that a web server is a typical server application. I suppose I should have been more specific; I was referring to business applications -- ERP, DSS, HR, A/R, A/P, Inventory Management, etc -- the type of applications that businesses pay money to have developed, because they need them to run their businesses. These sorts of applications are rarely I/O bound, because they typically have no more than hundreds of users and spend most of their time waiting on database queries and user input. With this sort of application, the use of multiple threads to manage code execution is perfectly sensible, and results in very clean and easy to maintain code. You'll note that I was not advocating writing webservers or databases in Java -- but backend applications that serve webservers and use databases are excellent targets for Java development.
Fair enough, I don't know what type of code you are writing. And if Java serves your needs than use it; I use it myself because it is frequently the best tool. And FWIW, I was thinking more along the lines of /dev/poll or kqueue() than select()
Bash c#,
Go back and read what I wrote -- I didn't claim that Java was as fast as C++ for *everything* -- just the sorts of applications I develop: business applications. Of course, if you continue to think of Java (and the JVM) as strictly an interpreted language, you're not keeping up with what the JIT compilers are doing these days, and you're completely ignoring the native-code compilers for Java.
Well, you said 'typical server applications' and you naturally think that means the kind you write, while I think it means the kind I write. I defer to you in determining what best suits your needs. I just get tired of people suggesting the JIT compilers or even native code producing Java compilers render C++, my personal fave, obsolete. Mostly just boosterism on my part.
Sorry for spelling: late, drunk.
Friends don't let friends post drunk.
The posting I can deal with. I just wish my friends would adopt some sort of stance on fat chicks.
Benchmarks have shown that for typical server applications, modern JVM/JIT combinations are as fast as unoptimized C++ code.
.net, etc, all you want. But don't tell me an interpreted language with garbage collection and bullshit container classes and 1970's i/o is as fast as c++. I write networking code in both languages, and i'm not buying it. Sorry for spelling: late, drunk.
load of crap. they may come close to being speed-equivalent if you write c++ constrained to the language features java supports. That is, you use heterogeneous container classes and downcast all the time, don't use templates or multiple inheritance, use threads where there is no good reason to, and whip expensive exceptions around like you're getting a commision on them.
Oh, and typical server applications?!? If you're writing server applications using a one thread-per-server-scheme, you are nowhere, and you are if you're writing java, cuz it doesn't know from multiplexing, at least in 1.3.
Bash c#,
use and distribution == plural a direct threat == single The elided infinitive 'to be' demands number agreement.
I may be misremembering this, but when I checked out Postgres it lacked outer joins and I couldn't live without them.
I work on Sun boxes all day long, because Sun's advertising and salesreps convinced someone in my company, who wouldn't know a line of code from a line of Goethe, that our company would be better off paying for Solaris than running Linux.
That's a problem, and not one that's going to go away while Sun looks to make a profit.
All the king's horses and all the king's men (1 experienced Oracle DBA and 1 good Linux sysadmin) couldn't get DBD::Oracle working with 8i on Intel redHat 6.0 Linux at my shop. .so libs in a dir and it would load every .so file in the disk except any one in the oracle tree (Forget the error now, sorry). The kluging that followed was horrific.
The make test failed loading a shared library, so I wrote a simple c prog just to load all the